The Impact of Legal Cannabis on Driving Safety

Does legalizing recreational cannabis and opening cannabis lounges make the roads less safe? It’s a loaded question.

Renee Consorte
Cannabis Explorations
11 min readJun 13, 2021


Photo by Aphiwat Chuangchoem on Pexels

Cannabis legalization efforts have been picking up steam in the U.S., with New York and New Jersey being the latest states to legalize cannabis for recreational use. According to an April 2021 report by the Pew Research Center, an increasing number of U.S. residents are proponents of ending marijuana prohibition, with nine-in-ten Americans supporting some form of legalization, either medical or recreational.

Many individuals who support legal recreational use, both marijuana advocates and regular Joes alike, view it as a step towards curtailing some of the damage that the harsh prison sentencing laws for nonviolent drug offenses has done and continues to do. These cruel and excessive legal punishments have disproportionately affected people of color, and the new legislation in New York has emphasized racial justice.

Other reasons that Americans support recreational cannabis, according to a 2019 Gallup Poll which surveyed the most common arguments U.S. adults had for and against legalization, include the plant’s perceived medicinal benefits, supporting individual freedoms, the opportunity for states to bring in tax revenue, and the fact that enforcing prohibition is expensive and resource-heavy for law enforcement.

It’s clear that legalizing marijuana federally is where we’re eventually headed as a country, but with this turning of the tides comes waves of concerns. The 2019 Gallup Poll said the most common reason cited by those who oppose marijuana legalization is the potential for an increase in the number of traffic accidents caused by drivers under the influence of cannabis. A whopping 79% of legalization opponents were worried about impaired drivers.

This worry may be magnified with the authorization of social consumption sites; places where adults can legally consume cannabis outside of their homes. Currently, only Colorado, Alaska, and very newly Nevada have statewide laws permitting and regulating these establishments. There are a few municipalities within other states such as West Hollywood, California which allow for the operation of these sites as well.

Policymakers around the country have run into significant hurdles when attempting to navigate legislation for social use. A proposed social use bill in Oregon back in 2017 died on the committee floor, as lawmakers could not reconcile its parameters with the statewide ban on smoking in the workplace. Public use laws often call into question states’ clean air policies and indoor smoking bans.

The defense for social use sites hinges on the issue that some people in states where marijuana is legal for recreational use, either residents or tourists, have nowhere to legally consume marijuana. Often, these states only allow marijuana consumption in private residences, putting a damper on the plans of tourists who come to the state specifically to purchase and consume it.

In New York, the new cannabis laws prohibit smoking it anywhere that it’s illegal to smoke cigarettes. This means that individuals who live in public housing don’t have anywhere they can legally smoke marijuana. This is a major reason why New York lawmakers are looking to allow cannabis lounges in the state.

Potential benefits of cannabis lounges aside, permitting social use begs the question: how pressing are the concerns about impaired driving? Will opportunities for social use make the roads less safe?

Is the risk as high as some claim?

With all the various reports and articles proliferated by opponents of legalization that show shockingly high levels of car accidents attributed to cannabis use, I can certainly understand what is fueling the fear of a large influx of intoxicated drivers.

The fact that high driving isn’t always taken seriously doesn’t quell anyone’s concerns either. A 2019 survey found that almost half of the participants who used cannabis felt it was safe to drive while high. Stoned driving has become a bit of a joke in certain circles; I’m sure many of us are familiar with the Tommy Chong bit where he jokes about getting pulled over by a caution light.

However, many studies that show alarming numbers of high drivers crashing their cars may be a bit biased.

A 2013 review of the literature on cannabis and driving risk, published in the journal Clinical Chemistry, discussed how early epidemiologic studies evaluating whether cannabis increases crash risk “didn’t provide strong evidence of cannabis causality”. These studies considered individuals who had only the non-psychoactive THC metabolite THC-COOH in their blood to be part of “cannabis-exposed groups”. Detecting this metabolite alone does not necessarily imply that an individual is intoxicated.

THC, the main intoxicant in cannabis, and its metabolites can be detectable in trace amounts in the blood for days after they are consumed, sometimes even weeks in heavy users. The effects of cannabis last only hours, so just because someone has a minuscule amount of a THC metabolite in their system doesn’t mean they’re actively intoxicated.

Cannabis’s culpability in traffic accidents is still not fully understood. It is the most common illicit drug detected in the bodies of drivers killed or injured in motor vehicle accidents, but that doesn’t necessarily mean that THC intoxication was responsible for all these crashes. Two reasons why it’s detected more is because it’s the most widely used illicit drug, and it’s also the most readily detectable drug in toxicological tests, even after the effects have worn off.

Also, many drivers who cause crashes have multiple drugs in their system at once. In 2016, 43.6% of fatally injured drivers tested positive for drugs, and over half of those drivers were positive for two or more drugs. Some reports lump cannabis-only drivers in with drivers who were also under the influence of alcohol or other drugs, muddying the causality further.

Though there are considerable issues with many studies falsely attributing some traffic fatalities to cannabis use, this doesn’t mean that THC doesn’t impair driving skills, or that it’s perfectly safe to drive high.

So…does cannabis increase crash risk?

A large study carried out by the National Highway Traffic Safety Administration (NHTSA) found that after controlling for demographic factors such as age and gender, and for blood alcohol level, those who tested positive for marijuana alone were not significantly more likely to be involved in a crash than a sober person.

Attempts to study the relationship between smoking marijuana and car accident risk has produced some conflicting results. A 2019 meta-analysis of 13 culpability studies found a small increase in overall crash risk. Other meta-analyses have had similar findings, yet a study of a sample of moderately injured drivers treated in the hospital after a crash found that those who tested positive for THC were not more likely to be responsible for a crash than those who had no THC in their systems, after adjusting for demographic factors.

Tests used to judge what functions are affected from cannabis use have shown that being high may cause impairment in virtually “every performance area that can be reasonably connected with safe driving of a vehicle”. These include motor coordination, tracking an object with the eyes and other visual functions, and perception of time and speed.

However, despite these concerning results, in actual driving tests high people only tend to show modest impairments. One possible explanation for the better-than-expected performance is that high people tend to overestimate how impaired they are and attempt to compensate. They drive slower and leave a greater distance between their car and the one in front of them. Drunk people, on the other hand, often underestimate how drunk they are and make far riskier decisions than they would sober, to devastating results.

Not all impairments can be compensated for though. Slowed decision making is a particularly significant deficit associated with cannabis that is hard to consciously make up for. Cannabis intoxication may also affect an individual’s ability to multitask or perform complex tasks, such as following directions from a GPS while trying to avoid potholes on the road.

Road tests which involved the participants actually driving in traffic demonstrated a relationship between dosage and lack of coordination. The higher the dosage of THC, the more the participant was prone to lane weaving. Those with the highest doses of THC had a comparable amount of difficulty staying in their lane to someone with a BAC of 0.05.

In addition, mixing alcohol and marijuana has been consistently shown to have an additive effect, so each drug increases the impairing effect of the other. This means that getting behind the wheel while crossfaded is especially dangerous.

While people who have consumed THC are generally thought to be less dangerous drivers than drunk people, and some studies attribute more traffic deaths to cannabis than is truly the case, it is generally accepted that stoned drivers are still more likely to make mistakes on the road than their sober counterparts.

The few studies addressing whether recreational states have seen an uptick in car crashes have not produced a definitive answer either.

One 2019 study found that Washington, Oregon, and Colorado all saw a temporary increase in the number of traffic deaths in the year following recreational legalization, by an average of one additional death per million residents. The trend was also seen in nine additional states either bordering on or near the legal states, including Kansas, California, and Nevada.

The temporary increase, which was followed by a downward trend and eventual normalization of the number of fatalities, could potentially be attributed to an influx of new weed smokers who were inexperienced with the drug, study co-author Tyler Lane told The Verge. This would explain why the increase was seen in both the recreational states and the states surrounding them, due to people crossing the border to purchase and consume cannabis.

A different study, done by the American Journal of Public Health, found no significant change in crash rates in legal states three years after legalization. Another study by the Insurance Institute for Highway Safety (IIHS) and the Highway Loss Data Institute (HLDI) reported a 6% increase in car crashes in 4 recreational states, but David Harkey, president of IIHS and HLDI, clarified in an interview with The Washington Post that more research is warranted to determine whether marijuana truly caused the accidents.

How is stoned driving prosecuted?

Another aspect of the situation which further unsettles both law enforcement and concerned citizens is the perceived difficulties of prosecuting stoned drivers. While there’s the well-established breathalyzer test that provides officers with undeniable evidence of alcohol intoxication, an equivalent does not exist for cannabis. Also, standard field sobriety tests verified to catch alcohol intoxication often aren’t the best at targeting stoned drivers.

Currently, the process of evaluating the status of someone suspected to be high relies mainly on subjective observation by an arresting officer and, at times, a Drug Recognition Expert (DRE), the latter being an officer with special training to recognize the signs of drug-induced impairment.

Six states have established legal limits for blood levels of THC, ranging from 1–5 ng/dL, so-called per se laws. Colorado has a similar rule, as drivers who have THC levels at or above 5 ng/dL in their blood are assumed to be impaired, but drivers who are charged with a DUI on this basis are able to contest their level of impairment despite their lab values.

But there is little evidence correlating a set amount of THC in the body to a consistent level of impairment across all cannabis users. An individual can be highly impaired with a small amount of THC in their blood, or quite functional with a level above the per se standard. THC is stored in fatty tissue, and frequent users who have a good amount of the chemical in their bodies may have a lingering low level of THC in their bloodstream.

“Low THC levels […] in blood can result from relatively recent use (e.g., smoking within 1–3 hours) when some slight or even moderate impairment is likely to be present, or it can result from chronic use when no recent ingestion has occurred and no impairment is present,” a report by the NHTSA stated.

Paul Armentano, Deputy Director of the National Organization for the Reform of Marijuana Laws (NORML), explained that the process of arresting someone suspected to be driving under the influence of drugs still has ways of gathering evidence of impairment, even without per se standards or an equivalent to the breathalyzer test.

“There would have to be tangible, demonstrable evidence of impaired driving, which is typically what leads to the pull-over in the first place,” he said. Behaviors such as lane weaving, running a stop sign, speeding, etc. are all cause for an officer to stop someone and to consider the possibility of impairment.

All other observations made by the officer, and the results of any field sobriety tests the officer conducts, as well as the 12-step protocol performed by a DRE if needed, contribute to the “totality of evidence”, as Armentano put it, that can serve to convict an intoxicated driver. This is the same process that is undertaken to investigate a driver intoxicated on other impairing drugs like opioids.

Since cannabis is so common and only becoming more so, Armentano is an advocate for adjusting field sobriety tests to better attune them to the effects of THC. Tests that look at things like short term memory recall and time perception could become a more normal part of an officer’s repertoire and help to identify people who are intoxicated from THC.

I think some people won’t ever be happy unless a test is developed that is an equivalent to a breathalyzer for empirically judging the level of THC in a person’s system. However, THC, and most other drugs for that matter, are not processed in the body in the same way that alcohol is, and there may never be a test that exactly mimics a breathalyzer for cannabis in terms of accuracy, efficiency, and speed.

How can we work to minimize intoxicated driving?

Armentano emphasized the importance of bolstering efforts to educate the public on the dangers of drugged driving.

“I think that any legalization or liberalization of marijuana laws […] should go part and parcel with public education efforts and campaigns to make people aware of the fact that it remains a criminal violation to drive under the influence of any impairing substance, including marijuana. And that the use of marijuana […] can influence a number of the skills that are necessary for safe driving,” he said.

He pointed out that increased awareness of the consequences of driving drunk, and the guidance provided through these education efforts on how to consume alcohol responsibly, has led to less and less people drinking and driving.

NHTSA data for 2019 showed that alcohol-caused car crash fatalities accounted for 28% of all fatal traffic accidents. Although drunk driving is still prevalent, this percentage was the lowest proportion of accidents for any year on record, since 1985. The data shows a steady decline in the number of fatal drunk driving accidents year over year.

When asked about the prospect of cannabis lounges, Armentano said that though “there is an argument to be made” for them opening up in certain jurisdictions, particularly those that are pedestrian-friendly tourist destinations such as Las Vegas, they’re unlikely to be nearly as widespread as bars that serve alcohol for the foreseeable future. They present significant regulatory hurdles that are difficult to overcome.

In terms of safety for those who visit these lounges, Armentano asserts that they could abide by some of the safety precautions that bar patrons should follow, but with some key differences of course.

Anyone who smokes marijuana should wait at least a couple hours before they even consider getting behind the wheel, or better yet appoint a designated driver or hire a ride.

It’s harder to predict how long a person will be intoxicated for from weed, unlike alcohol which has a pretty standard rate of elimination from the body. There is no clear established length of time for how long THC affects any particular person, and if someone is new to smoking or plans to smoke a lot in a short period of time, they should probably just not plan on driving at all.

My hope is that education efforts and revamped sobriety testing can assuage the doubts of those who worry about the safety of recreational cannabis.

If I could leave you with one final message: don’t smoke and drive.



Renee Consorte
Cannabis Explorations

I write about psychology and mental health, trying to understand why we act and feel the way we do.